A5103265463- Topic Modeling
- Natural Language Processing Techniques
- Advanced Text Analysis Techniques
- Physical Unclonable Functions (PUFs) and Hardware Security
- Integrated Circuits and Semiconductor Failure Analysis
- Software Engineering Research
- Random Matrices and Applications
- Spam and Phishing Detection
- Advanced Memory and Neural Computing
- Geochemistry and Elemental Analysis
- Algebraic structures and combinatorial models
- Advanced Combinatorial Mathematics
- Remote-Sensing Image Classification
- Atmospheric and Environmental Gas Dynamics
- Advanced Graph Neural Networks
- Geochemistry and Geologic Mapping
- Text and Document Classification Technologies
- Security and Verification in Computing
- Stock Market Forecasting Methods
- Recycling and utilization of industrial and municipal waste in materials production
- Coal and Its By-products
- Sentiment Analysis and Opinion Mining
Kunming University of Science and Technology
2025
Shandong Institute of Automation
2021-2024
Chinese Academy of Sciences
2018-2024
Nanjing University of Posts and Telecommunications
2023
University of Chinese Academy of Sciences
2021-2022
Research Center for Eco-Environmental Sciences
2022
Institute of Automation
2018-2021
Institute of Remote Sensing and Digital Earth
2017
We present an event extraction framework to detect mentions and extract events from the document-level financial news. Up now, methods based on supervised learning paradigm gain highest performance in public datasets (such as ACE2005, KBP2015). These heavily depend manually labeled training data. However, particular areas, such financial, medical judicial domains, there is no enough data due high cost of labeling process. Moreover, most current focus extracting one sentence, but usually...
Traditional approaches to the task of ACE event detection primarily regard multiple events in one sentence as independent ones and recognize them separately by using sentence-level information. However, are usually interdependent information is often insufficient resolve ambiguities for some types events. This paper proposes a novel framework dubbed Hierarchical Bias Tagging Networks with Gated Multi-level Attention Mechanisms (HBTNGMA) solve two problems simultaneously. Firstly, we propose...
Hang Yang, Dianbo Sui, Yubo Chen, Kang Liu, Jun Zhao, Taifeng Wang. Proceedings of the 59th Annual Meeting Association for Computational Linguistics and 11th International Joint Conference on Natural Language Processing (Volume 1: Long Papers). 2021.
We report the evolution of abundance, morphology, chemical species, and element fingerprints magnetic particulate matter during its emission process in thermal power plants.
Event information is usually scattered across multiple sentences within a document. The local sentence-level event extractors often yield many noisy role filler extractions in the absence of broader view document-level context. Filtering spurious and aggregating document remains challenging problem. Following observation that has several relevant regions densely populated with fillers, we build graphs candidate enriched by sentential embeddings as nodes, use graph attention networks to...
Most existing event extraction works mainly focus on extracting events from one sentence. However, in real-world applications, arguments of may scatter across sentences and multiple co-occur document. Thus, these scenarios require document-level (DEE), which aims to extract their a Previous cast DEE as two-step paradigm: sentence-level (SEE) fusion. this paradigm lacks integrating information for SEE suffers the inherent limitations error propagation. In article, we propose multi-turn...
Multi-sentence argument linking aims at detecting implicit event arguments across sentences, which is indispensable when textual events span multiple sentences in a document. Previous studies suffer from the inherent limitations of error propagation and lack explicit modeling local non-local interactions event. In this paper, we propose an event-aware hierarchical encoder for multi-sentence linking. Specifically, introduce to explicitly capture global Furthermore, auxiliary task predict...
In this paper, a high security encryption circuit based on ring oscillator PUF and secure scan chain is proposed. The composed of three parts: self-reference oscillator, test vectors input decision module circuit. key protection divided into two stages. the generation period, unique identification information chip generated by oscillator. transmission phase, XOR value correct obfuscation linear feedback shift register are chain, which further obfuscates data. Compared with conventional...
Scan chain insertion is a widely used technique in design for testability (DFT), which brings high fault coverage, good controllability and observability to chip testing. But the encryption chip, attacker can use scan carry out side channel attack on leak encrypted data chip. In order ensure security of scholars this field have done lot research. They exposed attackers against AES hardware algorithm methods, put forward series corresponding countermeasures, such as complex DFT structure,...
Since hyperspectral remote sensing (HRS) came in the middle 1980s, a number of imagers (e.g., EO-1 Hyperion) have been developed all over world. China, as one pioneers HRS technology development, also has active and contributed significantly to community. This paper updates recent advances future plans Chinese developments spaceborne imaging spectroscopy. Particularly, two powerful civilian micro/nano-hyperspectral mini-satellites (Spark01 Spark02) well China's first Carbon monitoring...